WG K6100 LV Supply Question

WindDrake

Well-known member
Joined
Apr 3, 2011
Messages
3,009
Reaction score
58
Location
Round Lake, Illinois
Hey folks.

Got my hands on a Tempest recently with a blown up LV section.

Common sense tells me to replace the LV section with an LV2000 and follow it around to make sure the HV section, etc is okay.

My coworker, however, has a different idea. He's got a couple of fairly beefy 25v linear power supplies that he wants to try and replace the entire +/-25v section with, completely eliminating the chassis NPN/PNP's and LV section with modern hardware.

Problem is, we can't figure out what the LV section is actually making, max current wise. The transistors are specced out to 10A, but with most things being built at ~50% overhead, we're wondering if a 5A supply will cut it. Nevermind we need to spec up proper fuses, etc.

Anybody know exactly or thereabouts what those supplies are capable of making under normal use?

Thanks!

Edit:

This thread has morphed into a WG 6100 repair log of sorts, once you get past the silly linear power supply questions.
 
Last edited:
Whoops! Realized this a bit ago as we were going over the monitor's manual.

The manual says 50V at 150 Watts maximum power consumption at max deflection. That means (About!) 3A. As it's two supplies, it's actually 25v*2, so that puts us at 3A/Rail.

We ended up fixing the deflection board (blown Q100/Q101 and R100/R101.) but we're still keeping this in mind for later.

Problem is, it just smoked something in the HV. Going to have to keep at it!
 
I think it's fairly uncommon for a failing LV to take out anything in the HV cage. Have you checked all the header pins where the boards connect up? It's very possible that it's just a bad connection. Are you getting neck glow?
 
Rebuild the HV section. Make dang sure the 555 is working. If there is smoke coming from around the little transfomer turn off ASAP.. If you fry the transformer,its a pain to find another.
 
Yep, everything in the LV's working, and getting neck glow.

Q906 is the culprit here (final drive before flyback in the boost section), dead short inside.

Checked the transistors/resistors in the HV, everything checks out.

Considering the game played for ~15mins before exploding Q100 (whoops! Subbed a 1/4 amp NPN, should have been a 1/2 Amp) it's likely the under-rated transistor was current-limiting the HV and preventing the blowout. Once we replaced it with a proper 1/2 Amp NPN, that's when Q906 decided it was done living.

Got the transistor on order. Guess we'll find out when it arrives. :)
 
Hm.

Pulled Q906, not short. Didn't realize that B>E was only a resistor and then the secondary side of the HV Transformer. Whoops!

So, with that out of the way, we figured out where the magic smoke came from. Q100 exploded (PNP) and R100 was toasted open.

R101 got hot at some point, but is still 3.9ohms. Q101 seems fine.

So, we replaced those, but we're still dealing with something pulling HARD on the positive supply, enough to roast these transistors over and over again.

Been through the HV board looking for off-rated resistors, caps with high IR/Shorts, and shorted diodes/transistors, but we're not having any luck. The chassis transistors also seem fine.

Any 6100 pros want to weigh in on this one?
 
Did you recap the HV section? Ive had this happen when on un rebuilt HV boards. There is an upgraded LV transistor (that Bobs kit will Downgrade..) I think its an NTE-50. its a TO-220 vs a to-92.
 
Haven't recapped the HV yet, no. With things metering decent, didn't think it'd be needed.

My coworker is going to raid his parts bins, otherwise we'll probably order a cap kit for the K6100.

I'm just stumped on the whole R/Q100 issue. It just seems really weird that it's getting pulled down on hard enough to nuke the transistor over and over again.

Edit: One thing I just remembered. The spot killer lit up when it originally let smoke out. Previously, that'd never happened before.

Guess I'll start looking at deflection, but I'm not seeing anything out of the ordinary there, either.
 
Last edited:
See if it will run with the HV unplugged. That will narrow it down to which board has the problem.

Usually when I troubleshoot these, I unplug everything from the deflection board. Once I get it stable, then I plug in the main power supply. Then each of the deflection outputs. Lastly, the HV.

All my meltdowns have either been:

1. Bad TO-3 or bad TO-3 mounting wafer.
2. Something on the HV board that goes away with bobs kit..;-)
3. 555 timer. (This is a nightmare)

Havent seen a flyback, HOT or that little guy hanging on the HV cage ever go bad.
 
I think you're right on the money with a deflection issue.

For giggles, we replaced Q100/R100, and tried it again, figuring that maybe last time, in our excitement (two experienced but giddy EE's does not make for reliable connector checking), that we hooked up something incorrectly.

HV came up, got neck glow, and we got a picture.

But we only got a picture on the top half of the screen. Aha!

And of course, the PNP transistor in the LV supply melted down again.

This pretty much points a big screaming finger to the X Deflection transistors. My coworker is digging up replacements for them, as he's pretty sure he has working subs for both the chassis transistors and the deflection board transistors.

We'll figure this thing out one way or another. :)
 
Did some more work on this.

Q100/101 both bad. Again. R100 open.

D105 shorted.

Replaced blown parts, hooked the LV board up with absolutely everything disconnected EXCEPT P100. Getting +/- 28v now.

Going to try slowly plugging things in at break here at work and see how it goes. Hopefully this way we'll narrow down what's causing all the drama.

Edit:

Progress! Got everything except HV and X/Y plugged in, nothing decided to quit.

Plugged in Y, no issues. Plugged in X, R100 got hot. Disconnected X and connected Y's transistor pair to X. No drama. This tells us the PNP (Negative Supply) chassis transistor is bad, as it's pulling excessive base current.

Swapping parts, will mess with it more after work. :)
 
Last edited:
Well, swapped X's PNP (Negative Side) transistor, same result.

Then realized one of the leads to Y's PNP was broken at the socket. Oops.

So, with either the X or Y transistor set plugged into X or Y amps, R100 gets hot.

Didn't get much time to mess with it, but we're going to try swapping X and Power transistor sets, to see if possibly the negative side chassis transistor for power can't handle the load.

May have to look into caps, also, but admittedly we're both a little stumped now. The supply is a rock solid 27.1vDC +/- when deflection is unplugged. We've got everything else hooked up EXCEPT the two HV connectors.
 
Hm, more progress.

Using a 25w 30ohm sandbar resistor, we test-loaded the positive and negative supplies.

They drop to ~15vDC, obviously incorrect.

Which means the damned thing has a regulation issue.

Going to study the schematic more. This one's got us headscratching.
 
Progress!

Checked out the supply again, this time at the 4-Diode Bridge Rectifier.

Under load, the bridge actually dropped about 2 volts. This is with a freakin ~0.83 Amp draw!

So, out went the four diodes, and we patched in a salvaged monolithic bridge rectifier from a dead 500w PC power supply.

Suddenly, the positive side of the supply can now take the load. Negative, however, still folds up.

So, we connected X-Deflection's transistor pair to J100, to see how it's chassis transistors fared.

Negative held, positive folded! It's the damned chassis transistors! They aren't driving to the rails, so the PNP/NPN's upstream are getting excessive collector current! Aha!

With this in mind, we did the same to Y Deflection.. And both supplies were stable!

So, that leaves us with a dead PNP in power, and a dead NPN in X Deflection. Previously, the PNP in X-Deflection was changed out with the NTE equivalent due to finding it was bad during unloaded power supply testing.

This means X totally failed on us at some point, and the PNP in power failed as well.

Got replacements at Frys, and we'll bolt it all up tommorow and see how it goes. We'll get this damned thing going yet. :)

Either way, LV2000 ordered today as well. Not screwing around with this unprotected nightmare LV circuit anymore. :)
 
Replaced the PNP in X and used that pair to run the power supply. Now stays stable using Y as the load. However, it does seem to drop about two volts, whereas the positive supply drops around a volt. Could be just the nature of the game playing and the demand on the negative, not sure yet without a complete load.

Replaced the NPN in the power transistor pair, and the negative side is still unstable, which means there's either something up with the new NTE285 transistor we used, or there's a harness issue.

Something from LV let smoke out, too. Probably the TO92 in the negative supply.

This thing is a beast. Ugh.

Edit: Checked harness with meter. 0.04 ohms resistance from plug to transistor socket. It's not the wiring, has to be the transistor. Damnit.

Should be getting that LV2000 Fri/Mon, so if we can't handle the retarded LV by then, I'm sure that'll help a lot.
 
Last edited:
More lunchbreak updates on this one.

The NPN in the negative side of the LV blew, just as we figured. Replaced it. Everything else was okay.

Loading both sides of the supply at the same time, we used the X transistors to run the supply. 25.3v solid. Same for the Y.

Retested the ones for power supply? Negative still folds up to ~18v... And positive also folds up.

That means the NTE285 we bought is definitely bad, and we'll need an NTE284 as well.

Progress is nice, but it's getting annoying to keep coming full-circle back to these damned chassis transistors.
 
Back at it.

Replaced the NPN and PNP in the power supply chassis transistors.

Still getting +/- folding up.

Plugged X's set into the power section, used Y as a load. Works fine.

Something is up here. We're going to go through the sockets and clean them up to see if it's a contact/current issue. This is beginning to get stupid.
 
Last edited:
Thought about this one more today.

Using the X Transistors, the power supply would hold steady, so what's different?

Harness tension.

Re-examined the J100 connector. Cold solder! D'oh! :( Considering all the work we did on this thing, I'm surprised two experienced techs missed this. Ah well.

Reflowed the entire Deflection board.

By god, it's alive!

Had to do some yoke turning and convergence, but it's about spot on. Got an improper sub in the negative side of X (bottom of screen) that we're swapping back to the original, then it's on to size and such to get the screen looking good again.

The 4700uf 50v line filters both had extremely high IR, so a pair of 1200 50v are in their place TEMPORARILY until the Digikey order gets in. It's holding without too much ripple, but it's causing deflection weirdness at the edges of the screen (foldback). This is evident the most when you go down the tube at the end of a level.

Nice to get this beast running, finally.
 
Back
Top Bottom